Relaxin 2 (H2 relaxin, RLN2) as a member of the insulin superfamily is a 2-chain peptide exhibiting, on the genetic level, the typical B-C-A chain prohormone structure, arranged from N- to C-terminus. Other members of this superfamily, encoded by 7 genes in human, are the relaxin genes RLN 1, RLN3, and the insulin-like peptide genes INSL3, INSL4, INSL5, and INSL6. The overall sequence homology between members of this family is low; nevertheless, phylogenetic analysis indicates that these genes have evolved from the RLN3 ancestral gene (Hsu, S. Y. (2003); Wilkinson, T. N. et al. (2005)). The mature protein has a molecular weight of approximately 6000 Da and is the product of an enzymatic cleavage of the prohormone catalyzed by the Prohormone-Convertase 1 (PC1) and 2 (PC2) (Hudson P. et al. (1983)). The resulting A- and B-chains are joined by two intermolecular cysteine bridges; the A-chain exhibits an additional intramolecular disulfide bond.
Relaxin initiates pleiotropic effects through multiple pathways on a variety of cell types. It confers its activity by binding to the class I (rhodopsin like) G-protein-coupled receptor termed LGR7 (leucine-rich G protein-coupled receptor 7) also named RXFP1 (relaxin family peptide 1 receptor), and with significantly lower affinity to LRG8/RXFP2 (relaxin family peptide 2 receptor) (Kong R C et al. (2010) Mol Cell Endocrinol. 320:1-15). Within the Relaxin molecule, an amino acid motif in the B-chain (Arg-X-X-X-Arg-X-X-Ile/Val-X) (SEQ ID NO: 162) (Schwabe and Büllesbach (2007) Adv Exp Med Biol. 612:14-25 and Büllesbach and Schwabe J Biol Chem. 2000 Nov. 10; 275(45):35276-80) is conserved in all of the Relaxin peptides and is crucial for the interaction of these peptides with the corresponding receptor. Binding of Relaxin to LGR7/RXFP1 leads to activation of adenylate cyclase and to an increase of the second messenger molecule cAMP. Via this mechanism, Relaxin 2 for example mediates the release of atrial natriuretic peptide in rat hearts (Toth, M. et al. (1996)). A positive inotropic effect of Relaxin 2 on rat atrial myocytes has also been shown (Piedras-Renteria, E. S. et al. (1997)). Other signal transduction molecules which are activated by the Relaxin/LGR7 complex are the phosphoinositide-3 kinase, tyrosine kinases, and phosphodiesterases (Bartsch, O. et al. (2001), Bartsch, O. et al. (2004)). Additional signal transduction pathways activated by this system include the nitric oxide (NO) pathway leading to increased levels of cyclic GMP in rat and guinea-pig hearts (Bani-Sacchi, T. et al. (1995)).
Relaxin acts as a pleiotropic hormone (Dschietzig T. et al. (2006)) possessing biological activity on organs such as lung, kidney, brain, and heart. A strong antifibrotic and vasodilator activity of Relaxin is most notably responsible for the positive effects obtained with this peptide in various animal disease models as well as in clinical studies (McGuane J. T. et al. (2005)). RLN2 has multiple beneficial actions in the cardiovascular system under pathological conditions. It maintains tissue homeostasis and protects the injured myocardium during various pathophysiological processes. It exhibits prominent vasodilatory effects, e.g. affecting flow and vasodilation in rodent coronary arteries (Nistri, S. et al. (2003)) and in the vascular beds of other organs. In spontaneously hypertensive rats RLN2 lowered blood pressure, an effect mediated by increased NO production.
A cardioprotective activity of Relaxin 2 has been evaluated in different animal models such as guinea pig, rat and pig (Perna A. M. et al. (2005), Bani, D. et al. (1998)). RLN2 ameliorates myocardial injury, inflammatory cell infiltration and subsequent fibrosis, thereby alleviating severe ventricular dysfunction (Zhang J. et al. (2005)).
Relaxin 2 exhibits strong antifibrotic activity. In injured tissues, fibroblast activation and proliferation causes increased collagen production and interstitial fibrosis. Fibrosis in the heart is increased by biomechanical overload, and influences ventricular dysfunction, remodeling, and arrhythmogenesis. In animal models, continuous infusion of Relaxin 2 inhibits or even reverses cardiac dysfunction caused by cardiomyopathy, hypertension, isoprenaline-induced cardiac toxicity, diabetic cardiomyopathy and myocardial infarction. This inhibition of fibrogenesis or reversal of established fibrosis can reduce ventricular stiffening and improve diastolic function. Notably, although Relaxin 2 reduces aberrant collagen accumulation, it does not affect basal collagen content in healthy tissues, highlighting its safety for therapeutic use.
Relaxin 2 has been tested in several clinical studies as a pleiotropic vasodilator for the treatment of patients with acute heart failure with very promising outcome. In these studies, Relaxin 2 was associated with favourable relief of dyspnoea and other clinical outcomes (Teerlink J. R. et al. (2009), Metra M. et al. (2010))
Due to the limited in-vivo half life of Relaxin, treatment of patients has to be repeated every 14 to 21 days, whereby compound administration has to be performed as a continuous infusion for at least 48 hours.
Furthermore, Relaxin 2 may also be useful in the treatment of diseases such as pancreatitis, inflammation-related diseases like rheumatoid arthritis, and cancer (Cosen-Binker L. I. et al. (2006) Santora K. Et al. (2007)) or scleroderma, pulmonary, renal, and hepatic fibrosis (Bennett R G. (2009)). Relaxin 2 reduces xenograft tumour growth of human MDA-MB-231 breast cancer cells (Radestock Y, Hoang-Vu C, Hombach-Klonisch S. (2008) Breast Cancer Res. 10:R71).
The synthesis of Relaxin 2 by chemical methods is difficult. Due to the low solubility of the B-chain and the requirement for the laborious, specific introduction of cysteine bridges between A and B-chains, yields of active peptide obtained by these methods are extremely low (Barlos K. K. et al. (2010)). Alternatively, recombinant expression of Relaxin 2 can be performed. To allow efficient cleavage of the prepro-peptide during post-translational modifications and the secretion of mature and biological active peptides, expression host cells are routinely co-transfected with expression constructs encoding the Prohormone-Convertase 1 and/or 2 (Park J. I. et al. (2008)). Nevertheless, the endoproteolytic processing efficiency of prepro-peptides in heterologous cells often limits the production of bioactive molecules significantly (Shaw J. A. et al. (2002)).
Therefore, it would be of great advantage to generate a Relaxin molecule which independent of endoproteolytic processing mediated by specific proteases exhibits full biological activity and can be produced in significant yields using heterologous expression systems.
For human Insulin, single-chain variants have been generated in which an uncleavable polypeptide connects the insulin B-chain with the insulin A-chain (Rajpal G. et al. (2009)). For these variants, endoproteolytic processing is dispensable.
Surprisingly, we identified a Relaxin variant in which the orientation of the two active chains, designated as A chain and B chain, are exchanged and the cleavable C chain is substituted by linker peptide. As shown in FIG. 1, instead of the genetically determined orientation of the single chains encoding Relaxin, namely B chain-C chain-A chain, the orientation of the chains of the modified molecule is: A chain-linker peptide-B chain. The resulting molecule exhibits full biological activity, independent of any endoproteolytic processing. This new single-chain Relaxin variant provided by the invention thus solves the problem of low expression yields or the requirement of co-transfection with a processing protease.
The half-life of intravenously administrated Relaxin 2 in humans is less than 10 minutes (Dschietzig T. et al. (2009)). As a consequence, in clinical trials Relaxin 2 has to be administered continuously over 48 h. Therefore, the improvement of the biological half life of Relaxin could be of great advantage.
Improving biological half life can either be performed by chemical modification such as PEGylation or HESylation of the polypeptide of interest, introduction of additional, non-natural N-glycosylation sites or by genetically fusing this polypeptide with other molecules such as the immunoglobulin Fc fragment of antibodies, transferrin, albumin, binding modules that bind in-vivo to other molecules mediating longer half-life, or other proteins, respectively. This invention provides single-chain Relaxin variants fused to the Fc part of antibodies with improved half-life. Surprisingly, these variants show biological activity in the range of the wild-type Relaxin.